Speech Recognition for Latvian
Collection
This collection contains open-source ASR models that we have fine-tuned for Latvian. • 7 items • Updated • 1
How to use AiLab-IMCS-UL/whisper-medium-lv-late with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("automatic-speech-recognition", model="AiLab-IMCS-UL/whisper-medium-lv-late") # Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("AiLab-IMCS-UL/whisper-medium-lv-late")
model = AutoModelForSpeechSeq2Seq.from_pretrained("AiLab-IMCS-UL/whisper-medium-lv-late")This model is a fine-tuned version of openai/whisper-medium on more than 20 hours of diverse Latvian broadcast data.
Test WER: 15.9%